muapi-3d-logo-animation

Pass

Audited by Gen Agent Trust Hub on Sep 30, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The material_style user input is interpolated into prompts for image and video generation tools. Although this represents an ingestion point for untrusted data, the risk is negligible as the target models are specialized for visual generation and lack general-purpose execution capabilities.\n
  • Ingestion points: material_style input in SKILL.md.\n
  • Boundary markers: None.\n
  • Capability inventory: muapi CLI tool and curl commands to the api.muapi.ai endpoint.\n
  • Sanitization: None.\n- [EXTERNAL_DOWNLOADS]: The skill interacts with the vendor's API at api.muapi.ai using curl for logo processing. This represents standard communication with the service provider's infrastructure.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 30, 2026, 07:17 AM